Learning to foucault

[skybadger]

Hi all 

I'm starting to learn this Foucault testing lark for testing ATM mirrors. The purpose is to able to grind and polish mirrors on my mirror o matic machine. I have a 6" ground to polishing stage and want to do a 12" f/5   for a travel scope and a 16" f/4 for a back yard visual scope. The latter o acquired part finished, hence wanting to get fully into the whole mirror grinding experience. 

I built/rebuilt a moving source slitless  Foucault tester and I'm using a known good 12" f/6 mirror to test against to learn the ropes. 

I've got texereau and a Foucault testing handbook as my manuals. I'm  looking for advice on how to improve contrast and repeatability. This is because I'm finding it {too} bright in the camera but fainter and low contrast to the eye and difficult to equalise the bulls eye and more so to repeat it to take measurements against.

Similarly I have real problems distinguishing the axial focus point.  Is this other people's experience ?

How can I improve contrast ?

How do you come reliably back to the same equal zones at the same measurement point in the mirror ?

Does the knife edge look rotated or not central? I suspect I need to drop the height a smidge to centralise the reflection but it's not easy!

The source  led is ground flat down to the zone where light is created in the led, so has a natural diffuser and is about .5mm across (?) The knife lies across the centre as close as I can estimate. I considered using a pinhole but there were some CN testers producing beautiful images which didn't even grind the led down or use a pinhole  so I'm not sure of the value of that. I guess I can test the use of one but I'll struggle to put half a 200um  hole across the knife. 

Some pictures are below.

The tester in the foreground and the mirror in the distance. The tripod carries a coarse xy adjuster and on top of that is the led source with single vertical knife on left/right micrometer and fore/back Vernier . On the hole where the eye goes I've put a small usb board camera and can slide it out on a dovetail. The camera connects to my phone for the pictures hence the lack of complex controls.  The blue cast in the pictures is the camera led I'll have to remove. The image sequence moves from outside focus towards the mirror to on axis focus.  Knife from right, moving source, so appears illuminated from left. 

 

Mirror in its cell and scope box in the distance. 

 

 

[Astrobits]

The problem with the Foucault  test is that the focus of the side zones is different and not on the optical axis. If you draw the light beams from a parabola out on paper you will see that the focus of, say, the left side 70% zone is on the right of the optical axis (OA) and the right side zone focus is on the left of the OA. With your knife edge on the OA you are cutting the beams away from their focus points. This doesn't pose a problem with large F/ ratios but becomes increasingly obvious as the F/ ratio gets smaller. To improve your accuracy with fast mirrors you will need to measure both X and Y positions of each focal point of each zone. These will lie on the Caustic curve.

Nigel

[bosun21]

2 hours ago, skybadger said:

and want to do a 12" f/5   for a travel scope

A 12” travel scope is too big for my travels.

[skybadger]

8 minutes ago, bosun21 said:

A 12” travel scope is too big for my travels.

I plan for it to be quite light..bar the weight of the mirror.

[skybadger]

2 hours ago, Astrobits said:

The problem with the Foucault  test is that the focus of the side zones is different and not on the optical axis. If you draw the light beams from a parabola out on paper you will see that the focus of, say, the left side 70% zone is on the right of the optical axis (OA) and the right side zone focus is on the left of the OA. With your knife edge on the OA you are cutting the beams away from their focus points. This doesn't pose a problem with large F/ ratios but becomes increasingly obvious as the F/ ratio gets smaller. To improve your accuracy with fast mirrors you will need to measure both X and Y positions of each focal point of each zone. These will lie on the Caustic curve.

Nigel

How do I finesse getting my tripod on the OA so that I can move backwards and forwards along the OA and know I'm sufficiently perpendicular to the mirror so my left and right knife measurements are comparable ? Otherwise I'll have to find the centre each time and take an estimate. 

In my head those annuli/zones from the parabola are focusing along the OA and I'm finding where each comes to a focus to measure the parabola. 

I've got to get to being able to do that reliably yet. Then I'll worry. 

Contrast and repeatability ? Necessity for a pinhole ?

[Astrobits]

I would not think that using a tripod will give you the delicate movements required. I would put the xy platform onto a flat surface so you can move it fractions of a mm at a time.

Nigel

[skybadger]

There is that. It's surprisingly stable but I haven't got a big bench I can dedicate to this.the image itself is stable, it just needs improving.

[fwm891]

I think you have too. bigger separation between the source and the return beam. When I made that tester I had less than 10mm between them. Large gaps will loose you contrast and make readings difficult.

Edited February 20, 2023 by fwm891
Text added

[davidc135]

Plenty of astigmatism coming from somewhere.  David

[Astrobits]

I agree with Francis.

If the camera is in the black disc on top of the circuit board then the separation between that and the source is far too large. You will need them to be as close as possible especially when you start figuring faster mirrors. Ideally the source and viewing point should be on the same axis. this can be accomplished with a beam splitter cube. When I figured my 16" F/4.7 mirror a separation of about 20mm showed significant astigmatism caused by that separation.

Nigel

[kbrown]

I don't know a lot about this but just wondering whether your test mirror is aluminized/silvered or bare glass. I'd imagine you'd typically test bare glass and testing a finished mirror will look a lot different in terms of brightness/contrast?

[skybadger]

That's a good point about the mirror being silvered.

Thanks. 

The separation between source and camera is about 15mm. I moved the source from horizontal to in the  vertical plane to allow source and image to be at the same distance from the mirror but that has increased the separation a little..

Looking at the stellafane examples I didn't think I was pushing any distance boundaries here. 

I can always try it and see. 

Need to do some more machining. 

I'll polish out the 6".f/5 to see how that affects the image brightness. 

Cheers

 

 

 

[skybadger]

7 hours ago, davidc135 said:

Plenty of astigmatism coming from somewhere.  David

You can see it Davide but you'll have to point it out to me, if you wouldn't mind ? 

That mirror has been on the bench for a few weeks so may have slumped a bit. 

What am I looking for  in those pictures ?

[markse68]

I don't think glass would slump at room temperature would it Mike?

Mark

[skybadger]

I believe it will, there's slump and there's slump..  

The images are of a high quality f/6 mirror. Texereau and Bartel's talk of slump in edge supported mirrors creating astigmatism, especially in thin mirrors. I'm told there is astigmatism in these images. Just guessing as to the cause. One could be due to the distance between source and image. One could be due to edge support. 

Rotation of the mirror would tell the difference.

 

[Astrobits]

Mirrors don't slump. I am not aware of any reports that a telescope mirror has changed shape from being in a mirror cell with three edge clips and stored sideways for years. It has been claimed that the glass in very old church windows is thicker at the base than the top due to the glass flowing under gravity ( slumping? ) over hundreds of years, but I am not aware that it has been proved to be the case. Very old glass sheet making did not produce the consistent thickness that we see today so thickness variance could well be there from day 1.

However, mounting mirrors vertically on two thin posts can induce localised stress which can show up in testing but I have not seen that particular feature during my mirror making experience despite my mirrors being mounted on two small supports.

Astigmatism can easily be seen using a simple lens ( 25mm eyepiece = 10x loupe ) by examining the return image of your source inside and outside of focus ( move the knife edge out of the way ). Both should be identical. Any elongation of the image one way compared to the other indicates astigmatism. It can also be seen in the Foucault test with the shadows not being symmetrical both parallel and normal to the knife edge. However, if the astigmatic axes are parallel/normal to the knife edge then they may be much more difficult to see in the normal course of Foucault testing.

Coated mirrors are remarkably bright compared to unsilvered ones.

Nigel

[Peter Drew]

My 30" F4.1 mirror is only 35mm thick at the edge, made in the 1980's (thin mirrors are nothing new 🙂).  Astigmatism induced by most means of edge support was avoided by having a 3" aperture perforation which enabled the mirror to be centrally supported on a stub projecting from the backplate, no edge support at all.  Seems to work.    🙂 

[inFINNity Deck]

3 hours ago, Astrobits said:

Mirrors don't slump. I am not aware of any reports that a telescope mirror has changed shape from being in a mirror cell with three edge clips and stored sideways for years. It has been claimed that the glass in very old church windows is thicker at the base than the top due to the glass flowing under gravity ( slumping? ) over hundreds of years, but I am not aware that it has been proved to be the case. Very old glass sheet making did not produce the consistent thickness that we see today so thickness variance could well be there from day 1.

...

Nigel

According to Edgar Dutra Zanotto glass does not flow, see E.D. Zanotto, 'Do cathedral glasses flow?', in: American Journal of Physics, 66(5), (1997), pp.392-395.

His conclusion was:

Quote

As a result of the previous discussions, it can be concluded
that medieval and contemporary window glasses cannot flow
at room temperature in human time scales!

Nicolàs

[Tomatobro]

Re the LED light source

My method for a light source was to shine a stopped down lamp onto a small ball bearing . Its been years since I did a mirror test but I found it gave very good even illumination of the mirror in a darkened room.

[skybadger]

Sorry for the delayed response. 

Maybe the term slump is the wrong term. What is happening to cause mirrors mounted on edge to give astigmatism that's removed by rotating ?. I guess it's bending ?

I've checked out the mirror I matic. It still functions. So next step is to cast my lap and polish the 6" to a sphere. Next post should be after a week or so. 

[markse68]

Flex- it should spring back instantly when the pressure is removed I think

Mark

[davidc135]

On 21/02/2023 at 20:50, skybadger said:

You can see it Davide but you'll have to point it out to me, if you wouldn't mind ? 

That mirror has been on the bench for a few weeks so may have slumped a bit. 

What am I looking for  in those pictures ?

There should be a symmetry with the light and shadow effects. See the centre left image where light and shade curl around each other. With a 6'' mirror being tested I doubt that either the optics or the stand is producing the gross effect. More likely there is something wrong with the set up. There's a lot of glare in the photography. Perhaps it's coma rather than astigmatism.

Here's an 8.5'' f7.5 mirror before refiguring. The Foucault test picks up all the roughness easily but not the mirror's considerable astigmatism which had nothing to do with the stand support or light beam separations, the mirror being 40mm thick and long focus.

I used a precision 50micron pinhole and knife blade and substituted a high power ep to follow progress regards astig. The Foucault test is sensitive to subtle differences of shadow.

David

Edited March 10, 2023 by davidc135

[Chriske]

There is indeed some surface roughness but very easy to remove.
Astig is very easy to detect even with Foucault.

In the position the mirror is now in I don't see any astig at all. Shadow is coming nicely in from the right(iow horizontal as it should).
Now rotate the mirror on it's stand for about 45° and take another picture. If the shadows do come in horizontal again, so in your case from the right, there is no astig. BUT, if the shadows come in at an angle -not parallel with the movement of the knife- then you do have astig.

btw best is to place the knife at the 70% zone, so you have a far better idea of the surfaces shape.

[skybadger]

On 21/02/2023 at 23:52, Astrobits said:

I agree with Francis.

If the camera is in the black disc on top of the circuit board then the separation between that and the source is far too large. You will need them to be as close as possible especially when you start figuring faster mirrors. Ideally the source and viewing point should be on the same axis. this can be accomplished with a beam splitter cube. When I figured my 16" F/4.7 mirror a separation of about 20mm showed significant astigmatism caused by that separation.

Nigel

I read this again. The camera is in the centre of the plastic rectangle. The black round disk is the original microphone from the camera, tucked out of the way, so the actual separation is about 10mm less than it look and may be about 15mm overall. 

I'll see what I can do to reduce it. 

What effects should I expect I'f the source and sensor are not the same distance from the mirror ? 

[Astrobits]

12 hours ago, skybadger said:

I

 

What effects should I expect I'f the source and sensor are not the same distance from the mirror ? 

Spherical aberration. Although virtually undetectable if the difference is small

Nigel